Sami,

Remember, depending on the analysis, you want to have normal RESIDUALS and
not necessarily normal distributions of the raw data.  It is quite possible
that you have a normal distribution of your residuals.

If not, since you have flux, you could always use the true value for N20 at
each measurement, and not flux.  The true value will always be positive.

Cheers,

Jim

On 1/30/07, Sami Ullah <[EMAIL PROTECTED]> wrote:
>
> Hello Ecologers:
>
> I have a data set of N2O fluxes from forest soils in which some values ar=
e
> positive and some negative. As the distribution in non-normal, I want to
> log-transform the data. However, the log transformation command in SAS
> transform positive values and ignores the negative. Can anyone guide me o=
n
> how to log-transform the data using SAS or anyother software to be able t=
o
> take care of both the positive and negative values.
>
> Thanks in anticipation.
>
> Sami Ullah
> Department of Geography
> McGill University, Canada
> [EMAIL PROTECTED]
>



--=20
James J. Roper
Depto Zoologia,UFPR
Caixa Postal 19034
81531-990 Curitiba, Paran=E1, Brasil
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D
E-mail:                           [EMAIL PROTECTED]
Phone/Fone/Tel=E9fono:        55 41 33611764
celular:                               55 41 99870543
Casa:                                 55 41 33857249
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D
http://jjroper.googlepages.com/

Ecologia e Conserva=E7=E3o na UFPR
http://www.bio.ufpr.br/ecologia/
                   ---

Reply via email to